Improving the effectiveness of training in teaching computer graphics is the quality control of the acquisition of training modules, as well as the organization of the educational process using new information and communication technologies. Given this, the article provides a theoretical analysis of the methodology for compiling test questions from computer graphics and a number of related graphic disciplines. The connection of factor theory with the graphic sciences was revealed. As a result, the structure of a three-level test system of problems, based on specific formulas and their factors, has been studied to date. In addition, the student grading system for four categories was tested using a common automated software tool for questionnaire and test control. Based on the results, a mathematical statistical analysis was carried out and a range of variations of four categories was shown.